To: atm{at}shore.net
From: mdholm{at}telerama.com
Reply-To: mdholm{at}telerama.com


Tom Krajci wrote:

Based on several years of use of both mirrors, star testing, etc....it
seems that if PLOP gives you RMS of about 2e-6 or better...your mirror cell
should be 'good enough' in all but the most ideal of observing conditions. 
(Assuming you build and use the cell properly.)


This is about 1/250 wave, depending on your reference wavelength.  This is
similar to the level I suggest as being sufficient for any ATM purpose on
my mirror cell web page.  Others, including David Lewis have suggested
somewhat looser limits, perhaps up to 4e-6mm or 1/125 wave RMS.  In between
is probably OK for most purposes.  Worse than 1/125 or so is probably not a
good idea. Better than 1/250 or so is probably not worth the trouble, since
something else will be the limiting factor.

One exception to the last sentence: If something else in the situation
influences the cell's effectiveness, such as contruction errors or an
intolerance for focus shift, it may be necessary to go to a more
conservative design.  Jeff Anderson Lee's analysis of cell construction
error effects (now linked to on my page) demonstrates this effect.

The mirror cell web page is at http://www.telerama.com/~mdholm/  the links
to Jeff's work is on the Links page.

P.S. I am working on a computer program to help those who want to lay out
cell parts to 1mm or better accuracy and who do not have access to a
milling machine.  Perhaps, if the rest of life doesn't keep interfering,
I'll have something ready in a week or so.

Mark Holm
mdholm{at}telerama.com

--- BBBS/NT v4.01 Flag-4